This watchlist tracks patent filings on sleep tracking, breathing sensors, heart rate accuracy during motion, muscle-signal wristbands, and hand or head gesture controls from Apple, Meta, and Google. Together they show three companies building wearables that sense your state without taps, buttons, or manual logging.
63 filings · tracking since Apr 2026 · latest Sep 2026 · updates weekly
based on all tracked filings in this watchlist · refreshes every week
These filings are all fighting over the same prize: making body-worn devices read your muscles, heart, blood, and movement accurately enough to be useful in real life.
Samsung and Meta are filing the most ground here, with Samsung pushing sensor accuracy across rings, earbuds, and watches, while Meta is almost entirely locked onto wrist-worn muscle-reading technology.
a dated entry each week this watchlist moves · older entries stay archived
This week adds two filings: one from Sony on a screen that grows when you turn your wrist, one from Google on reading blood pressure through a fingertip. Both point to bodies as the new control surface.
This week's filing shows Google working on heart-rate sensors that adjust based on a person's skin tone. The focus is on making body-reading wearables work more accurately for more people.
This week's filing comes from Sony, which is exploring a way to detect signs of sleepwalking before a person even gets out of bed. The focus is on reading body signals during sleep to catch a problem early.
Samsung leads this week with three filings covering how a device knows you're wearing it, keeps sensors accurate, and lets its screen check your health. Sony and IBM each added one filing, covering cleaner body signals and brain-to-search connections.
All three new filings are from Google, and they all point in the same direction: using cameras, earbuds, and screen images to read what your body is doing. The focus this week is on pulling health signals from everyday devices people already wear or look at, rather than dedicated medical tools.
counts from tracked filings · focus read from each company’s own filings
| Company | Filings | Last 8 wks | Focus | Latest move |
|---|---|---|---|---|
| Samsung | 18 | 8 | Sensor accuracy across devices | Samsung Patents a Two-Sensor Method for Telling When You're Actually Wearing a Device |
| 12 | 8 | Earbud sensing and AI | Google Patents a Fingertip Blood Pressure Sensor Built Into Everyday Devices | |
| Meta | 12 | 3 | Wrist muscle-reading bands | Finger Muscle Signals Could Soon Control Your 3D Workspace, Meta Patent Shows |
| Apple | 12 | 0 | Health data and audio gestures | Apple Patents a System Where AirPods Announce Themselves From Their Charging Case |
| Sony | 7 | 4 | Body tracking and sensors | Sony Patents a Display That Expands When You Rotate Your Wrist |
| IBM | 1 | 1 | IBM Patents Technology That Links Brain Signals Directly to Search and Retrieval | |
| Microsoft | 1 | 0 | Skin contact detection | Wearable Tech Patent Detects How Skin Shifts Against a Sensor |
20 or more filings in the last 8 weeks · 6 to 19 · under 6
the fights inside the fight · each with its three newest filings · new filings join every week
Meta 9, Google 1, Sony 1
Several companies are filing patents that read electrical signals from your muscles to let you control devices without touching them. Meta is the most active here, with Google and Sony also exploring how arm and finger movements could replace buttons.
Samsung 6, Meta 1
A cluster of patents tackles the basic problem that a ring or wristband moves around and gives bad readings. Samsung and Meta are both filing ways to detect how a device is sitting on your body and correct for it.
Google 4, Samsung 4, Meta 2
Getting clean heart rate and blood data from skin is hard, and Samsung, Meta, Google, and Apple are all filing patents that attack the problem from different angles, from filtering out noise to checking readings in two steps.
Google 4, Apple 3, Samsung 1
Google, Apple, and Samsung are filing patents that turn earbuds into body sensors, detecting whether they are in your ear, reading your temperature, or picking up on your gestures.
Apple 4
Apple is filing patents focused on how health information gets organized, shared across devices, and delivered to users in a way that is actually useful.
Meta wants finger muscle signals to control a 3D workspace, no touching required.
Samsung filed to change video content in real time based on your body's reactions.
Sony wants to identify individual gamers purely by how their muscles move.
Samsung's smart ring patent lets you control other devices just by gesturing.
every tracked filing, month by month · counts are USPTO pre-grant publications, one per publication number (an application, not a granted patent)
US 2026/0272323 A1
Blood pressure measurement has been missing from the wearables roster; this filing shows Google moving to embed the sensing hardware directly into device surfaces rather than requiring a separate tool or sustained contact.
US 2026/0277311 A1
Among the group's gesture-control approaches, Sony's method pairs finger position with wrist rotation to dynamically resize screen content, letting users expand displays without lifting or repositioning their wrist.
US 2026/0262975 A1
Heart-rate accuracy during motion depends on optical sensors that work differently across skin tones. Google's method uses existing phone photos to auto-calibrate sensor light levels before measurement, skipping the manual tuning that current devices require.
US 2026/0260758 A1
Sleep tracking gets a prediction layer: Sony's system reads pre-sleep physiology to forecast sleepwalking risk, then deploys monitoring if the score runs high.
US 2026/0248452 A1
Distinguishing skin contact from false readings requires Samsung's dual-receiver approach to filter out ambient light noise that single sensors mistake for actual body signals.
US 2026/0248464 A1
Wearables here need to know when you're actually still enough for an accurate reading. Samsung's approach puts that judgment into the display itself, automating measurement only when body position and posture match what produces valid data.
US 2026/0252578 A1
Brain-signal indexing expands gesture control toward hands-free memory search. IBM's neural interface captures cognitive state alongside environmental context, enabling retrieval by mental activity rather than typed queries.
US 2026/0248433 A1
Heart rate accuracy during motion improves when multiple electrode points replace single-sensor contact, reducing noise from movement and sweat rather than relying on one wrist placement to filter interference.
US 2026/0252142 A1
Wearable optical sensors need optical isolation to read accurately; Samsung's coating approach builds that isolation into the housing rather than relying on the sensor alone.
US 2026/0240489 A1
Sound-bounce detection in the ear canal extends sleep tracking into jaw movement patterns, grinding and chewing, without adding sensors elsewhere on the body.
US 2026/0244277 A1
Pairing camera vision with wrist-motion sensors lets the system confirm gestures through two independent channels, reducing false commands from camera occlusion or ambiguous arm movements.
US 2026/0245723 A1
Camera-based screen capture bypasses the wearable-to-phone connection problem, letting users photograph any health display and have the data automatically parsed into their device instead of manual transcription.
US 2026/0236777 A1
Separating heart rhythm into slow and fast components lets the AI extract distinct medical signals instead of processing raw waveforms whole, mirroring how trained cardiologists read ECGs layer by layer.
US 2026/0236099 A1
Detecting finger muscle signals without visible hand movement lets users navigate AR interfaces while keeping their hands occupied or still, bypassing the current need for hand gestures or controllers.
US 2026/0232248 A1
Wrist placement shifts optical sensor readings by enough to throw off heart rate data, so Meta's system uses a reference sensor to calibrate position without user guesswork.
US 2026/0224120 A1
A grid of pressure sensors in a finger cuff maps arterial pressure changes across skin contact points, replacing the arm-wide inflation needed for traditional BP measurement.
US 2026/0227867 A1
Wearables translate body movement into on-screen cursor control via wireless link to the display, extending gesture recognition beyond handheld remotes into full-body input.
US 2026/0224394 A1
A flexible strap with an internal locking tab keeps pressure sensors stable during motion, solving the drift problem that degrades accuracy in wrist and finger-worn devices across the watchlist.
US 2026/0230672 A1
Swapping video content based on real-time heart rate and eye strain readings expands the watchlist beyond passive monitoring into active intervention, using wearable sensors to reshape what you watch mid-stream rather than simply logging vital signs.
US 2026/0215696 A1
The watchlist has focused on sensor accuracy during motion; this filing shows how a ring solves that through physical contact, keeping optical sensors pressed to skin despite finger movement, which matters for heart rate and blood oxygen readings.
US 2026/0219742 A1
Heart rate and blood oxygen readings from wearables require sensor orientation. Samsung's ring detects which way it sits on your finger to correct measurements automatically.
US 2026/0215706 A1
The heart-rate accuracy challenge extends beyond motion to signal strength itself. Meta's approach catches weak optical readings before they become bad data, automating the sensor recalibration that users now do manually or simply accept as failure.
US 2026/0219749 A1
A ring that requires deliberate touch before recognizing hand gestures solves the false-trigger problem that has plagued gesture-control wearables, moving hand and head gesture controls from noisy sensor data to intentional commands.
US 2026/0222751 A1
Acoustic reflections from the ear canal let earbuds confirm proper insertion and detect user adjustment, adding a passive fit-verification layer to wearables that already track heart rate and breathing during wear.
US 2026/0211496 A1
Gesture controls have relied on wrist motion or hand position so far; Samsung's groove-sensor ring shifts to finger contact itself as the input method, letting the ring detect when you press an adjacent finger against it.
US 2026/0211504 A1
Muscle-signal wristbands need electrode placement that captures distinct finger movements without crosstalk. Meta's patent maps how spacing and arrangement on the band surface improve signal separation between individual digits.
US 2026/0207139 A1
Embedding sensors into the band material itself solves a real durability problem: keeping muscle-signal electrodes stable against skin through normal wear, sweat, and movement instead of relying on surface contact that shifts.
US 2026/0211502 A1
Capacitive skin detection lets the device distinguish worn contact from removal or hand-off, adding biometric security to the wearables watchlist without requiring active user input like passwords or gestures.
US 2026/0207162 A1
Acoustic waves bouncing through tissue let the wristband measure core temperature rather than skin temperature, extending the watchlist from heart rate and breathing into metabolic sensing without requiring invasive probes.
US 2026/0211505 A1
Detecting hand gestures from muscle signals eliminates the need for physical contact with controllers or screens. This approach lets users navigate AR interfaces through natural hand movements read directly from the wrist.
US 2026/0198792 A1
Detecting low-light conditions lets wearables switch sensor modes automatically, improving heart rate accuracy when ambient light drops too low for reliable optical measurement.
US 2026/0199793 A1
Muscle-signal authentication extends the wristband sensor concept beyond gesture control into biometric identity verification, letting hardware recognize individual users from neuromuscular signatures alone.
US 2026/0195906 A1
Hand and head gesture controls need reliable skeletal maps. Sony's body-worn camera system skips the reflective dots and external rigs, building joint positions from cameras mounted directly on the wearer.
US 2026/0182849 A1
The muscle-signal and heart-rate tracking efforts have run into optical noise from device internals. Sony's light-steering approach isolates the detector from stray reflections to improve reading fidelity.
US 2026/0182916 A1
The watchlist has covered heart rate and breathing sensors; this filing extends body-reading into composition metrics by taking dual-point electrical measurements from a single wearable contact, moving beyond single-site approximations.
US 2026/0182906 A1
A pulse sensor using dual-wavelength light filters can measure melanin content alongside heart rate, potentially calibrating readings across different skin tones without requiring a separate sensor.
US 2026/0186532 A1
Dual optical sensors at different distances let Samsung filter ambient light and skin reflections separately, sharpening heart rate signals during the motion-heavy moments when wearables struggle most.
US 2026/0174340 A1
Combining muscle and heart sensors in one wristband requires solving signal interference, Meta's approach isolates the two measurement types electronically so they can operate simultaneously without corrupting each other's data.
US 2026/0181372 A1
Placing earbuds in their case triggers automatic pairing signals instead of requiring manual discovery, streamlining how wearables connect to phones and other devices.
US 2026/0175084 A1
A wearable that merges calorie intake with exercise targets in real time, turning the device into an active nutritional feedback loop rather than a passive tracker that logs data separately.
US 2026/0178613 A1
The sync problem cuts deeper than just collecting readings across devices, Apple's approach shows the real challenge is keeping conflicting data versions from corrupting each other when multiple wearables feed the same health metrics into one account.
US 2026/0178268 A1
Hearing voices in noisy settings requires burying yourself in menus. Apple's patent shortcuts Conversation Boost into the volume control, collapsing multiple taps into a single gesture so users can isolate speech without breaking focus on their surroundings.
US 2026/0169685 A1
Detecting hand and head gestures without a screen requires audio feedback so users know their movement registered, not just guessing whether a nod worked.
US 2026/0157640 A1
Telling muscle from fat requires more than a scale. Apple's camera method lets fitness apps swap generic weight numbers for actual body composition, so a runner and lifter doing the same workout get different calorie and recovery estimates.
US 2026/0164162 A1
Acoustic sensing inside the ear canal lets earbuds detect jaw and head movements without cameras or touch inputs, filling a control gap for hands-free, silent interaction in public or crowded settings.
US 2026/0162657 A1
A wearable that monitors speech patterns in real time could flag cognitive states, attention lapses, fatigue, stress, before the wearer notices them, extending health tracking from physiology into mental performance during daily activity.
US 2026/0161138 A1
The timeline so far tracks how wearables sense what's happening inside your body. This filing adds a different layer: using multiple wearables worn together to fix gaps in location tracking when GPS drops out in tunnels or dense urban areas.
US 2026/0156400 A1
Detecting acoustic leakage in real time lets earbuds self-correct their seal instead of forcing users to manually refit them during use. This moves wearable sensing from passive monitoring into active compensation.
US 2026/0151037 A1
Inferring core body temperature and breathing rate from facial thermal readings eliminates the need for separate sensors, moving vital sign monitoring from wrist bands and chest straps into the headset itself.
US 2026/0151089 A1
Skin temperature mapping tells the device whether it's flush against bare skin or sitting over fabric, letting it auto-correct readings that drift when the fit changes during activity.
US 2026/0151085 A1
A smartwatch that requires two consecutive abnormal readings before alerting you could prevent the anxiety spikes that make people distrust their wearables' health warnings.
US 2026/0155913 A1
A relay system between paired earbuds lets the stronger receiver retransmit packets to the weaker one, keeping calls stable when one bud loses direct connection to the phone.
US 2026/0155983 A1
The ecosystem angle: instead of each wearable collecting its own biometric data, one trusted device vouches for you across the others, reducing enrollment friction and keeping the primary phone as the security anchor.
US 2026/0151071 A1
A wristband that reads muscle signals needs its internal antenna to reliably reach every sensor around your wrist; this patent solves the wireless routing problem so signals from scattered electrode patches actually make it back to the processor.
US 2026/0153928 A1
The wearable timeline so far has focused on passive sensing, heart rate, breathing, sleep. This filing pivots to reading muscle signals for active control, turning your arm into an input device rather than just a measurement surface.
US 2026/0147407 A1
Head gestures join the tracking mix, but Google's two-step sensor method solves a practical problem the earlier patents sidestepped: filtering real nods from random movement noise.
US 2026/0142893 A1
A camera-based gesture system lets users control speaker volume and playback from across a room without touching a phone or speaking commands, extending wearable sensing beyond the body itself to hand movements in physical space.
US 2026/0137291 A1
A wearable that knows when its own sensor can actually see straight eliminates a major source of false readings during exercise, where noise from movement has always corrupted heart rate data.
US 2026/0133646 A1
The wearables watchlist has focused on what sensors can detect. This filing solves a basic signal-quality problem: eliminating electrical noise that corrupts biosensor readings in the first place.
US 2026/0133634 A1
Muscle-signal sensing requires filtering out electromagnetic noise from surrounding devices. Meta's shielding design isolates the wristband's sensors so they can detect faint muscle activity in ordinary environments instead of controlled labs.
US 2026/0119024 A1
Tracking sleep consistency requires knowing whether you actually stayed in bed when scheduled, not just whether your phone dimmed at the right time. Apple's patent automates the morning accounting of how often you hit your bedtime targets.
US 2026/0120824 A1
Apple's system filters health data through the lens of upcoming activities, surfacing relevant metrics rather than everything at once. This moves wearable tracking from passive collection toward active decision support.
US 2026/0118675 A1
A face-worn device with light sensors could measure breathing depth and pattern without contact, filling a gap where current wearables rely on chest placement or audio detection.
No. These are patent filings, which describe research directions and possible designs, not confirmed products or launch dates. Some ideas in this watchlist may show up in future devices, others may never ship. The value of tracking them is seeing where each company is investing engineering attention, not predicting a release calendar.
Getting sensors to work reliably outside a lab is the recurring theme. Apple's motion-gated heart rate system and light-based breathing sensor, and Meta's shielding and grounding work for muscle-reading wristbands, all address the same issue: real bodies move, sweat, and generate electrical noise that clean lab data doesn't account for.
Meta's filings concentrate almost entirely on one device type, a muscle-signal wristband, and solve internal engineering problems like grounding and shielding. Apple spreads its filings across phones, headsets, and watches, patenting sleep mode switching, breathing sensors, and heart rate accuracy rather than a single wearable form.
Several patents, from Apple and Google, aim to let people control devices without touching a screen, using hand motions in the air or a nod and shake of the head. That points toward wearables and earbuds that respond to your body's movement directly, cutting out taps and app switching.
Want this weekly breakdown for a company we don't cover? Patentlyze Pro →